ModelsCompareBest forBenchmarksStatusPricingAPI
Compare / head-to-head

DeepSeek V4 ProvsGemini 3.5 Flash

Gemini 3.5 Flash leads 5 of 7 shared benchmarks. DeepSeek V4 Pro is 8x cheaper per token. Gemini 3.5 Flash has the larger context window (1.0M tokens).

Benchmarks from cited public sources; pricing from official pages; status from official provider feeds.

Shared benchmarks
2 – 5
Gemini 3.5 Flash leads
Cheaper per token
DeepSeek V4 Pro
8x cheaper, input + output
Larger context
Gemini 3.5 Flash
1.0M tokens
Provider uptime (30d)
100% · 100%
DeepSeek · Google
DeepSeek V4 Pro
DeepSeek · released 2026-08-13
text
Context
1M
Max out
384K
Input /1M
$0.435
Output /1M
$0.87
Cached /1M
$0.004
Scores
11 · 9 core
Gemini 3.5 Flash
Google · released 2026-05-19
textvisionaudio
Context
1.0M
Max out
66K
Input /1M
$1.5
Output /1M
$9
Cached /1M
$0.15
Scores
15 · 9 core
Quality

Benchmark matrix

BenchmarkDeepSeek V4 ProGemini 3.5 FlashΔEdge
Reported by both · 7
GPQA Diamond90.9% epoch92.8% epoch-1.9 ptGemini 3.5 Flash
AIME 202696.7% matharena95% matharena+1.7 ptDeepSeek V4 Pro
FrontierMath Tier 4 v2 (Epoch AI run)2.4% epoch26.8% epoch-24.4 ptGemini 3.5 Flash
FrontierMath Tiers 1-3 v2 (Epoch AI run)45.3% epoch62.8% epoch-17.5 ptGemini 3.5 Flash
OTIS Mock AIME 2024-2025 (Epoch AI run)96.7% epoch95.6% epoch+1.1 ptDeepSeek V4 Pro
SimpleQA Verified47% epoch66.2% epoch-19.2 ptGemini 3.5 Flash
SWE-bench Verified (Epoch AI run)77.6% epoch79.3% epoch-1.7 ptGemini 3.5 Flash
Only DeepSeek V4 Pro reports · 4
LMArena Elo1450.6 not reported
DeepSWE v1.162.7% not reported
Humanity's Last Exam (with tools)60% not reported
Terminal-Bench 2.187.9% not reported
Only Gemini 3.5 Flash reports · 8
ARC-AGI-2not reported72.1%
GDPVal-AAnot reported1656
Humanity's Last Exam (full set, text + MM)not reported40.2%
MCP Atlasnot reported83.6%
MMMU-Pronot reported83.6%
OSWorld-Verifiednot reported78.4%
SWE-Bench Pro (Public)not reported55.1%
Terminal-bench 2.1not reported76.2%
Scores tagged "epoch" or "matharena" are independent runs, used only where the lab has not published its own; ⚠ marks rows MathArena flags as released after the competition. Higher is better on every row. Δ is DeepSeek V4 Pro minus Gemini 3.5 Flash in the benchmark's own unit. "Not reported" means the lab has not published that figure; it is not a zero. ↗ opens the source.
Specs & pricing

Side by side

SpecDeepSeek V4 ProGemini 3.5 FlashEdge
Context window1M tokens1.0M tokensGemini 3.5 Flash
Max output384K tokens66K tokensDeepSeek V4 Pro
Input price / 1M$0.435 $1.5 DeepSeek V4 Pro
Output price / 1M$0.87 $9 DeepSeek V4 Pro
Cached input / 1M$0.004 $0.15 DeepSeek V4 Pro
Input + output / 1M
Lower is cheaper. List prices; batch, tool and regional fees excluded.
$1.31$10.5DeepSeek V4 Pro
Modalitiestexttext · vision · audioGemini 3.5 Flash
Released2026-08-132026-05-19
Cited benchmark scores1115
Reliability

Provider status

All providers
More matchups

DeepSeek V4 Pro vs …

More matchups

Gemini 3.5 Flash vs …

Built by Respan
Which one wins on your data?

Public benchmarks are a starting point. Run DeepSeek V4 Pro and Gemini 3.5 Flash on your own prompts with Respan evals, or route to either through one gateway key with automatic failover.